Hi All,
I’m encountering a critical issue in QuickSight when applying an aggregate filter in my analysis. I’m trying to build an analysis/dashboard to track the login users in my org, and one of the charts I’m trying to build is a donut chart to reflect the distinct count of users who have not logged in during a specific timeframe, group by user profile. My logic is to group user data by a field in my dataset called “user_profile” and display the count of distinct “user_id” field with the filter aggregated sum of field “login_count” being 0. However, while the data is expected to show 17.86k users when correctly grouped by user_id, it incorrectly shows only 4 users across 2 user profiles when grouped by user_profile. This issue is persistent and replicable across all visualization formats (bar chart, table) and appears to be linked specifically to the aggregation filter I applied, I double checked the dataset and I’m sure there’re no NULL value in the dataset that could potentially impact the result.
Attached are the relevant screenshots, including my dataset, the donut chart, the fields and filter I applied to the donut chart, and the donut chart with right number 17.86k, which for testing purpose I grouped by user_id.
